Telemetry for Tornex using :telemetry.

Tornex.Telemetry defaults to using Logger to handle telemetry events, but this behavior can be customized by not attaching the default telemetry handler.

Telemetry Events

Tornex dispatches the following events through :telemetry:

• [:tornex, :api, :start]: Executed when there is an API call started against a Tornex.Query or a Tornex.SpecQuery.

  • Measurement: %{}
  • Metadata: %{resource: String.t, resource_id: integer, selections: [String.t()], user: pos_integer}

• [:tornex, :api, :finish]: Executed when the API call started against a Tornex.Query or a Tornex.SpecQuery has finished.

  • Measurement: %{latency: pos_integer}
  • Metadata: %{resource: String.t, resource_id: integer, selections: [String.t()], user: pos_integer}

• [:tornex, :api, :error]: Executed when an API call has failed.

  • Measurement: %{}
  • Metadata: %{}

• [:tornex, :api, :timeout]: Executed when an API call is over the timeout value and the API call has been ended.

  • Measurement: %{}
  • Metadata: %{}

• [:tornex, :bucket, :create]: Executed when a new Tornex.Scheduler.Bucket has been created for a user ID.

  • Measurement: %{}
  • Metadata: %{user_id: pos_integer, pid: pid}

• [:tornex, :bucket, :create_error]: Executed when a Tornex.Scheduler.Bucket could not be created due to some issue.

  • Measurement: %{}
  • Metadata: %{user_id: pos_integer, error: term}

• [:tornex, :bucket, :timeout]: Dispatched by a Tornex.Scheduler.Bucket when it has been empty for too long and has been killed.

  • Measurement: %{}
  • Metadata: %{pid: pid}

• [:tornex, :bucket, :enqueue]: Dispatched by a Tornex.Scheduler.Bucket when a Tornex.SpecQuery has been run against it or has been enqueued to it.

  • Measurement: %{}
  • Metadata: %{selections: [String.t], resource: String.t, optional(resource_id): integer, user: pos_integer}